Conventionally, the following type of technology is known. Specifically, lane boundary lines (referred to in the following as white lines) that define the traffic lane in which a host vehicle is traveling (referred to in the following as the host-vehicle lane) are identified based on captured images of a region ahead of the host vehicle. The traffic lane width (displacement-possible region) of the host-vehicle lane is then estimated based on the identified white lines, etc.
The traffic lane width which is thus estimated is used, together with the current position of the host vehicle, the velocity vector, and the vehicle width, etc., as information that is required for executing control (referred to in the following as intra-lane travel control) in the host-vehicle lane, with a known type of lane-departure warning and driving support, etc.
Furthermore technology has been proposed (see PTL 1) with such a type of intra-lane travel control, whereby when the white lines cannot be identified but the width of a preceding vehicle can be calculated, the traffic lane width is estimated based on the width of the preceding vehicle.